Alejandra Rodríguez

Automated knowledge-based analysis and classification of stellar spectra using fuzzy reasoning

Abstract This paper presents the application of artificial intelligence techniques to optical spectroscopy, a specific field of Astrophysics. We propose the analysis, design and implementation of an intelligent system for the analysis and classification of the low-resolution optical spectra of supergiant, giant and dwarf stars, with luminosity levels I, III and V, respectively. The developed system automatically and objectively collects the most important spectral features, and determines the temperature and luminosity of the stars according to the current standard system. The system development combines signal processing, expert systems and fuzzy logic techniques, and integrates them through the use of a relational database, which allows us to structure the collected astronomical data and to contrast the results of the different classification methods. As an additional research, we have designed and implemented several models of artificial neural networks, including them as an alternative method for the classification of spectra.

A comparative study of KBS, ANN and statistical clustering techniques for unattended stellar classification

The purpose of this work is to present a comparative analysis of knowledge-based systems, artificial neural networks and statistical clustering algorithms applied to the classification of low resolution stellar spectra. These techniques were used to classify a sample of approximately 258 optical spectra from public catalogues using the standard MK system. At present, we already dispose of a hybrid system that carries out this task, applying the most appropriate classification method to each spectrum with a success rate that is similar to that of human experts.

An Artificial Neural Network Approach to Automatic Classification of Stellar Spectra

This paper presents the design and implementation of several models of artificial neural networks for the automatic classification of low-resolution spectra of stars. In previous works, we have developed knowledge-based systems for the analysis of spectra. We shall now use these analysis methods to extract the most important spectral features, training the proposed neural networks with this numeric characterization. Although there are published works about neural networks applied to the classification problem, our final purpose is the integration of several artificial techniques in a unique hybrid system. In the development of such a system we have combined signal processing techniques, knowledge- based systems, fuzzy logic and artificial neural networks, integrating them by means of a relational database which allow us to structure the collected astronomical data and also contrast the results achieved with the different classification methods.

Automatic Classification of Optical Sources Candidates to Be in the post-AGB Stage

As a part of an ongoing program, we have carried out a survey searching for stars in their late stages of evolution (post-Asymptotic Giant Branch; post-AGB stars) by observing low-resolution spectra of about 200 objects, all of which need to be classified in the MK system. We present an intelligent system aimed to classify the spectra of post-AGB stars in the MK system in an automatic and objective way. For the development of the system we have combined signal-processing techniques with knowledge-based systems, which allow us to integrate in a unique system tools for analyzing and classifying stellar spectra. The integration of the processing techniques and the knowledge-based system is performed by means of a relational database, which includes symbolic and numerical information.

STARMIND: Automated Classification of Astronomical Data Based on an Hybrid Strategy

This paper describes the formulation and development of STARMIND, a hybrid system devoted to the automated classification of stellar spectra in the MK system. The MK system is an astronomical classification system used to cluster stars in morphological types based on stellar temperatures and luminosities. Our hybrid system is composed by a knowledge-based system that performs the first taxonomy in stellar types. A second-level system is based on Artificial Neural Networks and performs a more refined classification in stellar subtypes. Artificial Neural Networks were defined by selecting the optimal algorithms for training and architecture for each of the stellar spectra subtypes.

An intelligent system for the spectral classification of stars - artificial neural networks vs. statistical clustering techniques

This paper presents an intelligent system for the classification of low-resolution optical spectra of the stars in the current standard MK system. We propose a comparative analysis of two techniques, artificial neural networks and statistical clustering algorithms, applied to the spectral classification of a sample of approximately 258 optical spectra from public catalogues. We do not only intend to analyze the efficiency of these two approaches in the automatic classification of spectra; our final objective is the integration of several techniques in a unique intelligent hybrid system. This system is capable of applying the most appropriate classification method to each spectrum, which widely extends the research in the field of automatic classification.

Expert systems and artificial neural networks applied to stellar optical spectroscopy: a comparative analysis

This work presents a comparative study of two computational techniques - expert systems and artificial neural networks - applied to a specific field of astrophysics, the classification of the optical spectra of stars. We present a description of various expert systems and neural networks models, and the comparison of the results obtained by each technique individually and by a combination of both. We do not only intend to analyse the efficiency of these two approaches in the classification of stellar spectra; our final objective is the integration of several techniques in a unique hybrid system. This system will be capable of applying the most appropriate classification method to each spectrum, which widely opens the research in the field of automatic spectral classification.

An Artificial Neural Networks Approach to the Estimation of Physical Stellar Parameters

This paper presents an artificial neural networks approach to the estimation of effective stellar temperatures by means of optical spectroscopy.